enter image description here

There are serveral things I (yet) don't understand about the T-T-earthing system. The most important one is that I don't understand which way the the current takes in case one of the conductors (for example L1) touches the chassis of the device.

There are two options I can think of:

a) There is a flow of charge out of the earth at the first grounding, through the Resistance RB, through the coils of the transformator, through L1, through the chassis (at the location of the fault) , through RA, into the ground at the location of the grounding of the device.

There is not a closed current loop. Instead, ground just works like a really big capacitor (with a very big capacity), so draining charges out out of the earth, or putting it in, will not change the potential of the ground at all.

b) There is in fact a current loop in case of a faulty connection between L1 and the chassis of the device, starting at RB, through the coil of the transformator, through L1, through the Chassis, through RA and back to RB.

This would mean there actually is current flowing through the earth (which I find hard to believe since there are parts of the earth that have a very high resistivity.

Which of the two options I suggested is true? Or does it indeed not matter how to model this, because the values of RA and RB can account for the different ways of modelling this?

Edit: I come from another field (physics), and while I have a good understanding about electric potential an electrodynamics, I'm neither familiar with the concepts used in electro-engineering, nor with the terminology. That's where this question stems from.

  • \$\begingroup\$ The ground is a more like a resistor than a capacitor. \$\endgroup\$ Mar 21, 2019 at 22:01
  • \$\begingroup\$ @Jasen A resistor between what? Between the chassis and the star-point of the transformator? \$\endgroup\$ Mar 21, 2019 at 22:08
  • \$\begingroup\$ Grid Earth bonding is low R // big C so don’t try to short it to chassis due to ESR of big C. The C is water and the actual impedance for a step load covers a broader spectrum than a sine wave \$\endgroup\$ Mar 21, 2019 at 22:20
  • \$\begingroup\$ betqeen any earth connections. like that theoretical infinite resistor array. \$\endgroup\$ Mar 22, 2019 at 2:34

3 Answers 3


It's essentially option (b). The fault current travels through the Earth between the two grounding rods.

Parts of the Earth may have a high resistivity, but at the same time, it is a really fat conductor - about 12000 km at its widest!

But, the resistances where the earth rods meet the ground, modelled as RA and RB, may be quite substantial. This means that the fault current can be quite low, often not even enough to blow a fuse or trip a breaker. That is why a TT installation should have an RCD (or GFCI).


Answer: b) There is in fact a current loop thru ground.

The demands for low R require a depth with ground moisture such that the moisture dielectric gives a relatively low impedance limited by the ESR of the conductive salts in the earth and the dielectric moisture. This is often measured by the effective series resistance in Ohm-m.

Here are some stats for HV grid dynamic fault current vs required detection time and Resistivity for different soil moisture content.

enter image description here ref

Notice my RC estimates below, by coincidence, correspond to an RC time constant just shorter than the maximum response time to reduce Fault damage to the secondary and earth bond network, for a remote fault. As the fault gets closer to the source, these times reduce.


simulate this circuit – Schematic created using CircuitLab

In addition, thermal and mechanical stresses to the customer's ground grid and ground grid connections can increase the grid's resistance to ground and, at the same time, fault potentials. In order to prevent these problems from occurring, a ground grid assessment, utilizing field and utility updated data, should be carried out on a regular basis. This paper will illustrate a European Committee for Electrotechnical Standardization (CENELEC) approach to ground grid design, aimed to maximize the electrical safety under ground fault. In addition, case studies will be included, showing how high fault currents have damaged ground grids and what repairs are possible.



There will be fault currents through the earth, for this reason RCDs are almost always used to protect users on TT circuits.

  • \$\begingroup\$ I'm sorry, but this answer could apply to either of the options ( a) and b) ) that I asked about. Can you be more specific? \$\endgroup\$ Mar 21, 2019 at 22:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.